ESP Journal of Engineering & Technology Advancements |
© 2021 by ESP JETA |
Volume 1 Issue 1 |
Year of Publication : 2021 |
Authors : Dr. B. Sakthivel |
: 10.56472/25832646/ESP-V1I1P104 |
Dr. B. Sakthivel, 2021. "Low Power Design of Geometric Mean Filter Using GWO Pruning" ESP Journal of Engineering & Technology Advancements 1(1): 17-21.
Geometric mean filter is commonly used in image processing application to remove Gaussian noise. In filter stage requires more number addition and multiplication process .Pruning is an approximation technique used to achieve a low power processing .In this work, Grey wolf optimized (GWO) geometric mean filter designed for low power image processing application. To validate proposed technique in real time scenario, the images with Gaussian noise are considered and GWO-Pruned 16-bit multipliers are utilized to perform multiplication between 16-bit pixels. . The proposed method gives the highest PSNR and computation cost than conventional pruning technique. On an average, the proposed technique has improved the PSNR rate of quality of 8.213 % and achieves energy requirement reduction of 19.36 %, respectively.
[1] Xiao Chen & Cheng, 2014, “An imprecise algorithm for real-time compressed image and video transmission”, Proceedings of Sixth International Conference on Computer Communications and Networks.
[2] Liu, W, Chen, L, Wang, C, O’Neill, M & Lombardi, F 2014, “Inexact floating-point adder for dynamic image processing”, 14th IEEE International Conference on Nanotechnology.
[3] Merhav, N 1999, “Multiplication-free approximate algorithms for compressed-domain linear operations on images”, IEEE Transactions on Image Processing, vol. 8, no. 2, pp. 247-254.
[4] Suman, S, Hussin, F, Malik, S, Walter, N & Goh, L 2014, ‘Image enhancement using geometric mean filter and gamma correction for WCE images’, Lecture Notes in Computer Science, vol. 5, no. 2, pp. 276-283.
[5] Paim, G, Rocha, G, Da Costa, E & Bampi, S 2019, ‘Maximizing the power-efficiency of the approximate pruned modified rounded DCT exploiting approximate adder compressors’, 17th IEEE International New Circuits and Systems Conference (NEWCAS).
[6] Tu, J, Zhan, Y & Han, F 2010, “A neural network pruning method optimized with PSO algorithm”, Second International Conference on Computer Modeling and Simulation, pp. 246-259.
[7] Leon, V, Zervakis, G, Soudris, D & Pekmestzi, K 2018, ‘Approximate hybrid high radix encoding for energy-efficient inexact multipliers’, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 26, no. 3, pp. 421-430.
[8] Kim, S & Kim, Y 2016, “Energy-efficient hybrid adder design by using inexact lower bits adder”, IEEE Asia Pacific Conference on Circuits and Systems (APCCAS).
Filter, Geometric, Low Power